Mathematical Algorithm on Conventional Computerized Adaptive Testing

Suhardi, Iwan (2016) Mathematical Algorithm on Conventional Computerized Adaptive Testing. In: PROCEEDING OF 3 RD INTERNATIONAL CONFERENCE ON RESEARCH, IMPLEMENTATION AND EDUCATION OF MATHEMATICS AND SCIENCE, 16-17 Mei 2016, Yogyakarta.

	Text (Artikel Prosiding) ME-56.pdf Download (433kB)
	Text (Peer Review Artikel Prosiding) Prosiding Internasional - Mathematical Algorithm on Conventional Computerized Adaptive Testing.pdf Download (930kB)
	Text (Hasil Turnitin) TURNITIN PROSIDING 2.pdf Download (450kB)
	Text (Bukti Korespondensi) KORESPONDENSI 2 SERTIFIKAT ICRIEM.pdf Download (55kB)

Abstract

In education, it is important to determine the ability of the students on a subject. By knowing it, a teacher can take appropriate action to deliver good education. One method to determine the ability of the student is by performing adaptive testing. Adaptive testing is a testing method in which each examinee will be given a different set of questions based on the ability of each student. Thus, each examinee do not need to answer all questions. The items were selected with specific procedures based on the estimated level of ability of the students which reflected on their responses. Adaptive testing can be automated using a computer device, called a computerized adaptive testing (CAT). The CAT based on item response theory (IRT) with one parameter logistic model, two parameters, and 3 parameters. CAT has many advantages compared to other testing applications and has been shown to have high efficiency and reliability. CAT is also very capable to be developed further, both in terms of procedure and also the application. Eventhough there has a lot of references about CAT and its development, but in reality it is not easy to build a CAT application program as a whole. In this paper, the authors will share the experience in developing a conventional CAT application in a detailed mathematical algorithms and examples of results analysis obtained by the response answers from the students. This paper is expected to provide an initial basis for other developers and CAT researchers to build, develop and further improvement of the CAT program for large-scale applications.

Item Type:	Conference or Workshop Item (Paper)
Subjects:	FAKULTAS TEKNIK
Depositing User:	Arsad Bahri
Date Deposited:	26 Jun 2023 04:15
Last Modified:	26 Jun 2023 04:15
URI:	http://eprints.unm.ac.id/id/eprint/30220

Actions (login required)

View Item